Download Multiregional hypothesis explained

AMERICAN JOURNAL OF PHYSICAL ANTHROPOLOGY 112:129 –136 (2000)
Multiregional, Not Multiple Origins
MILFORD H. WOLPOFF,1* JOHN HAWKS,2 AND RACHEL CASPARI1
Paleoanthropology Laboratory, Department of Anthropology, University
of Michigan, Ann Arbor, MI 48109-1382
2
Department of Anthropology, University of Utah, Salt Lake City,
UT 84112
1
KEY WORDS
multiregional evolution; modern human origins;
polygenism; parallel evolution; independent origins; simultaneous evolution; treeness
ABSTRACT
Multiregional evolution is a model to account for the pattern
of human evolution in the Pleistocene. The underlying hypothesis is that a
worldwide network of genic exchanges, between evolving human populations
that continually divide and reticulate, provides a frame of population interconnections that allows both species-wide evolutionary change and local
distinctions and differentiation. “Multiregional” does not mean independent
multiple origins, ancient divergence of modern populations, simultaneous
appearance of adaptive characters in different regions, or parallel evolution.
A valid understanding of multiregional evolution would go a long way toward
reducing the modern human origins controversy. Am J Phys Anthropol 112:
129 –136, 2000. © 2000 Wiley-Liss, Inc.
The past decade has seen very significant
fossil (Duarte et al., 1999; Li and Etler,
1992; Lü, 1990; Pope, 1992) and genetic
(Awadalla et al., 1999; Bower, 1999; Harding et al., 1997; Harris and Hey, 1999; Hey,
1997; Loewe and Scherer, 1997; Relethford,
1998; Wise et al., 1997, 1998) discoveries
that directly address the modern human origins debate. Many believe they should have
resolved it. But the debate has not been
resolved. Several ideas about why the modern human origins debate continues unabated have been proposed, ranging from
the influence of strong feelings and exacerbating personal comments in books and
newspapers, to the contention that the debate cannot be resolved because the hypotheses are not sufficiently contradictory or
even are untestable (see Clark and Willermet, 1997; Howell, 1996; Smith and Harrold, 1997). We believe there is a much simpler explanation. For whatever reason, or
reasons (and we suspect there are more
than one), there has been a continued pattern of misinterpreting or incorrectly de©
2000 WILEY-LISS, INC.
scribing multiregional evolution. These misinterpretations follow a theme, in which
multiregional evolution is depicted as parallel or independent evolution (often involving simultaneous changes) in the different
inhabited regions of the world,1 and then is
rejected because such a scheme is unlikely.
There have been attempts to correct this
(Frayer et al., 1993; Relethford, 1995, 1999;
Smith, 1997; Wolpoff and Caspari, 1997).
There are historical reasons for this particular misinterpretation (Wolpoff and
Caspari, 1997). Here we do not discuss these
but rather focus on the literature of the past
*Correspondence to: Milford H. Wolpoff, Paleoanthropology
Laboratory, Department of Anthropology, University of Michigan, Ann Arbor, MI 48109-1382. E-mail: [email protected]
Received 6 July 1999; accepted 13 January 2000.
1
One of the reviews of this paper notes that “if you propose,
and then espouse, a hypothesis about modern human origins
and call it the ‘Multiregional Hypothesis,’ it is probably likely
that people will think . . . that your hypothesis involved ‘modern
humanness’ having more than one origin.” But apart from the
fact that multiregional evolution is a hypothesis about the pattern of human evolution, and not specifically about modern human origins (if there is such an event; see Wolpoff and Caspari,
1996), “multiregional” refers to nothing more or less than “many
places” and there is no reason to assume it means “many places
130
M.H. WOLPOFF ET AL.
few years. The papers we discuss reflect the
consistent consequences of interpreting
multiregional evolution to mean parallelism
and the simultaneous evolution of modernity. More than any other source could, they
demonstrate by example why it is that without a valid understanding of the multiregional model, it is not possible to examine
its consequences, test it, or make progress
toward resolving the modern human origins
debate. We present several examples from
genetics and paleoanthropology.
INTERPRETATIONS OF MULTIREGIONAL
EVOLUTION
Independent origin of Homo sapiens
in China
As part of their analysis of microsatellite
relationships between contemporary Chinese populations, Chu et al. (1998) discussed the multiregional evolution hypothesis and stated that it is incorrect because
“genetic evidence does not support an independent origin of Homo sapiens in China”
(Chu et al., 1998, p. 11763). Multiregional
evolution does not predict the “independent”
evolution of modern humans in China, or
anywhere else in the Old World. It evokes
diffusion across a network of genic exchanges, a mechanism that is the opposite
of independent evolution, to account for the
shared pattern of evolution across the human range combined with the presence of
some regional continuities in various areas.
The multiregional hypothesis is that a network of genic exchanges, promoted by but
not necessarily dependent on exogamy
rules,2 provides a frame of population interconnections that allows both species-wide
where the same things happen at the same time independently
of each other.”
2
Native Australian marriage preference is defined by descent
group, they must marry outside their clan, but they are not
required to marry outside their “tribe” (defined by dialect and
territory, these are the smallest breeding populations on the
continent according to Birdsell, 1958). Yet even without “tribal”
exogamy rules, the data of Tindale (1940) on Native Australian
marriages show that for each generation, on average 13% of
marriages were with a neighboring “tribe” and 1.6% were with
“tribes” whose borders were not adjoining (Lasker and Crews,
1996). Ignoring population movements and reticulations (which
would speed up the process), this is more than sufficient to
insure a much greater magnitude of genic exchange than the less
than one migrant per generation between regional populations
that the multiregional model requires to work (Relethford, 1999;
Relethford and Harpending, 1995).
evolutionary change and local distinctions
and differentiation. As the evolving human
populations continued to diverge and reticulate, this network has had several consequences. It encouraged geographic differentiation through isolation-by-distance for
neutral traits. For traits that were not neutral, it allowed advantageous features, promoted by selection, to spread everywhere
throughout humanity. Traits such as these
are widely shared, but differentiation in
other adaptive traits across this network
reflects adaptive variation, tempered by historic differences.3 Even when the distribution of the causes of adaptation are not clinal, there can be widespread gradations of
variation maintained by gene flow balanced
against selection differences or drift. Because of the key role played by genic exchanges in this model, multiregional evolution means that no human species,
subspecies, or race can have multiple “independent origins” in different regions. If genetic loci have evolved in the absence of
selection, as is often assumed for microsatellites such as those examined in Chu et al.
(1998), then multiregional evolution predicts there will be a pattern of isolation-bydistance and expects reticulate evolution
among populations.
However, Chu et al. (1998) provide no test
of whether the Chinese microsatellite data
fit an isolation-by-distance model, nor do
they assess the effects of population reticulation on the observed population structure.
Instead, they use a standard algorithm to fit
their data to a branching model, or tree, of
population relationships. Just as regression
may be fit to any pair of metric variates, any
matrix of population distances may be fit to
a tree. However, as in the case of regression,
we must evaluate a priori whether a tree is
an appropriate model for the population relationships. In the case of these microsatel-
3
A new source of selection can only modify what is already
present (Gould and Vrba, 1982), barring the unlikely occurrence
of a useful mutation at just that moment. Therefore, while
population differences in adaptive features reflect adaptive variation, the details of the differences may, in some cases, be found
in variation created by random genetic drift at an earlier time.
The same difference in selection does not necessarily cause the
same difference in the morphological response to it, so population history can be a key element in this process even before the
process has begun.
MULTIREGIONAL, NOT MULTIPLE ORIGINS
lite data, there are two obvious hypotheses
to explain the observed interpopulational
distances. First, we may believe that a series of population divergences, perhaps accompanied by founder events, led to the differences. If this were true, a tree would be
the appropriate model to fit to the data.
Second, we may believe that differences in
migration among the populations led to the
differences, in which case a tree would be
clearly inappropriate. Either of these two
hypotheses will result in a matrix of population differences, and the observation of
such a matrix is not sufficient to determine
which hypothesis is the correct one (Relethford, 1995). They may be distinguished,
however, by testing the goodness of fit of the
data to a tree model, measuring the “treeness”4 of the data. In this case, Chu et
al.(1998) do not report the treeness of the
microsatellite data. This is an important
omission, both because analysis of diverse
human genetic data demonstrates that treeness can be rejected for global samples of
living human groups (Templeton, 1998),
and because a high treeness is the only way
that distance data could refute an isolationby-distance model for humans in China.
The further conclusion that “modern humans originating in Africa constitute the
majority of the current gene pool in East
Asia” (Chu et al., 1998, p. 11766) is also
compatible with the multiregional hypothesis, given the stipulation, common to most
modern human origins explanations, that
until recently more people lived in Africa
than in other parts of the world (Relethford,
1999). In this respect, multiregional evolution and uniregional replacement have similar expectations, since Africa makes up half
or more of the inhabited land mass of the
world prior to 100,000 years ago. In contrast, multiregional evolution can easily be
disproved if it can be shown that all of the
ancestors of living humans at some discrete
time in the Middle or Late Pleistocene lived
in only one area of the world. If this were the
4
“Treeness” is exhibited when all of the endpoints on one side
of a split are equally related to all of the endpoints on the other
side. If there were migration and genic exchanges between
groups at the endpoints, then the relations would depend on the
pattern of migration and not on the split. In this case, equal
relationships would not be found.
131
case, then we should be able to trace the
ancestry of every human genetic locus to a
single population existing at some time in
the past million years. This testing is actively underway using genetic evidence, and
no such time has yet been found at which
every genetic locus resides in a single African population (Harris and Hey, 1999; Harding et al., 1997). However, such a test has
not been performed on microsatellite data,
and no such test is performed in this study.
Unique descent
The commentary accompanying Chu et al.
(1998) emphasized the inference that the
majority of East Asian genes came from Africa, and concluded, “This should help refute the claim that there is a continuity of
evolution from Homo erectus to modern humans in East Asia, as maintained by supporters of the multiregional hypothesis”
(Cavalli-Sforza, 1998, p. 11502). This is a
misrepresentation of the term “continuity”
as explicitly employed in the multiregional
model, by confusing the continuity of features with a claim of unique descent,5 and
as noted above, most modern human origins
explanations agree about where the majority of Pleistocene humans lived.
The Neandertal lynchpin
It is further stated (Cavalli-Sforza, 1998,
p. 11502) that “another stronghold of the
multiregional hypothesis was the transformation of Neanderthal into modern humans
in Europe, and also this has been falsified
by an analysis of DNA of the Neanderthal
par excellence.” Europe (especially Western
Europe) and the fate of the Neandertals
have never been a “stronghold” of the multiregional evolution hypothesis, which was
5
A claim of unique descent from ancient to modern Asian
populations would not be multiregional evolution, but polygenic
evolution. Regional continuity refers to the observation that very
common features persist in different regions for long periods of
time. It is not the claim that such features do not appear elsewhere; the genetic structure of the human species makes such a
possibility unlikely to the extreme. There may be uniqueness in
combinations of traits, but no single trait is likely to have been
unique in a particular part of the world, although it might
appear to be so because of the incomplete sampling provided by
the spotty human fossil record. Neither is it the claim that such
features persist for the entire period of habitation of a region, for
such a claim would disregard population replacements and extinctions, as well as the action of natural selection for some
features, all of which are usual evolutionary events.
132
M.H. WOLPOFF ET AL.
explicitly developed to account for the fossil
evidence in East Asia (Wolpoff et al., 1984).
The interpretation by Cavalli-Sforza
(1998) is based on the assumption that the
Neandertal’s distinct mtDNA lineage is a
separate biological lineage, reflecting a population (or paleodeme) that diverged from
humanity when the mtDNA lineage
branched, and subsequently evolved in parallel with humans. But mtDNA lineages are
not separate populations, and population
genetics demonstrates that the reported Neandertal mtDNA cannot falsify any relevant
evolutionary model for European origins
(Nordborg, 1998; Wolpoff, 1998). To date,
there is only one published partial sequence
of Neandertal mtDNA, and it has not been
compared with the mtDNA of its contemporaries, but instead only with that of living
humans. The problem is, as Nordborg
(1998) lays out, that only a very few ancestors of the world sample of human mtDNA
were present in contemporaries of the Feldhofer Neandertal. This means that the total
sample of mtDNA that could be compared is
very small. At the moment, the analysis
that has been done has demonstrated little
more than what has been accepted paleontologically for over 100 years: Neandertals
were biologically at the fringes of the range
of variation of living humanity. No Neandertals can be found today, although most of
their features remain, at differing frequencies (Frayer, 1992). Moreover, and perhaps
most relevantly, multiregionalism could be
a valid explanation for human evolution
even if every Neandertal became extinct
without issue. No human populations persist endlessly or continuously through time;
all either become extinct, or merge with
other populations.
An extreme theory of multiple origins
In another paper, multiregional evolution
was portrayed as an “extreme version of the
origin of modern humans (Homo sapiens sapiens) from Homo erectus . . . [with] multiple
origins, one in almost every continent” (Li et
al., 1999, p. 3796). There is a serious question about whether modernity can be defined in a way that would allow a statement
like this to be valid (Wolpoff and Caspari,
1997), and no multiregionalist has proposed
a subspecies taxonomy for modern humans,
but even more serious, and disturbing, is
the persistent contention that multiregionalism means multiple origins. The fundamental difference between multiregional
evolution and all replacement theories is
that multiregional evolution describes a
process within a single evolving species and
therefore is a reticulate model in which
there is branching, extinction, and merging
of populations. In contrast, replacement
theories, such as the single recent African
origin of modern humans, describe a process
of evolution by branching alone, in which
reticulations are impossible. The branching
process is assumed a priori and is the basis
of analyses to account for both the origin of
modern humans and the differentiation of
these humans into the different, widespread
populations found today.
It is correct to say that multiregional evolution and evolution by replacement are extremes, because there is no process that can
lie between them or be a compromise. But
authors who continue to regard the multiregional hypothesis as a “candelabra theory”
(Lewin, 1993; Seielstad et al., 1999) are simply incorrect. Ironically, it is the branching
replacement model that must be described
as a “candelabra” (Templeton, 1998), because branching is the only explanation this
theory offers for variation. In the replacementist view, morphological variation in the
Middle and Late Pleistocene is the result of
branching of different species, and genetic
variation among living humans is the result
of the branching of ancestral human populations. By assuming population variation
to be a reflection of the time since common
ancestry of diverging populations, replacementists are forced to accept parallel evolution to explain Middle and Late Pleistocene
changes across the human range, and complete population isolation to explain the observed worldwide pattern of genetic diversity in living humans.
Parallel independent evolution
The depiction of multiregional evolution
as a polygenic theory of multiple origins and
parallel evolution continues in paleontology
as well as in human genetics. In the most
recent paper by the two paleoanthropolo-
MULTIREGIONAL, NOT MULTIPLE ORIGINS
gists who disagree with multiregional evolution most strongly, we find the comment,
“Frayer (1992) also proposes, on the basis of
the persistence of traits across the Neandertal/modern boundary, a degree of continuity
indicating ‘some measure of genetic contribution of Neandertal to subsequent Homo
sapiens populations’ . . . this view no longer
means regional evolution” (Bräuer and
Stringer, 1997, p. 199). But given the reticulate nature of evolution within species, and
the ethnogenic modeling that is part of the
multiregional framework (Moore, 1994),
what else could regional evolution mean unless multiregional evolution is still being interpreted by these scholars as parallel independent evolution in different regions?
Elsewhere, Stringer and McKie (1996, p.
141) assert that “multiregionalism . . . holds
that our brain development is an event of
all-consuming global consequence towards
which humanity strived in unison for nearly
two million years. . . . to believe that humanity could be the product of a small, rapidly evolving African population who struck
it lucky in the evolution stakes is therefore
viewed as being worse than apostasy by
these people.” Gould (1994) writes, “Multiregionalism is awfully hard to fathom. Why
should populations throughout the world,
presumably living in different environments, under varying regimes of natural selection, all be moving on the same evolutionary pathway?” Howell (1996, p. 32) asserts,
“MRE requires that natural selection drive
African and African-derived hominin populations in Eurasia anagenetically and ineluctably toward the modern human condition. It has an almost omega-point
inevitability about it.”
Simultaneous appearance of modernity
as mutations accumulate because of
relaxed selection
Natural selection is not the only force
“posited” to drive simultaneous parallel evolution in these innovative depictions of multiregional evolution; there is also the absence of natural selection. Smith and
Harrold (1997), writers who do not particularly support the Eve theory, provide an
example which mistakes multiregional evolution for parallelism and uses the absence
133
of selection to account for it. The model they
describe as multiregional evolution is actually that of Brace (1991), i.e., “stages of human evolution.”6 Brace (1991, p. 52) asserts
that multiregional evolution is, in actuality,
no more than his PME (probable mutation
effect), when he contends, “The idea of the
emergence of modern human form gradually and simultaneously throughout the entire occupied world or ‘in situ continuity’ has
recently been rechristened ‘multiregional
evolution.’”7 According to Brace (1991), his
worldwide evolutionary stages are based on
a theory of parallel evolution that evokes
human culture as the common source of parallel trends that cause simultaneous
changes, and the accumulation of mutations
in the absence of selection8 as the mechanism that creates the parallelisms (culture
removes selection, in Brace’s view).
Simply put, Smith and Harrold (1997)
turn to this cultural explanation for common parallel evolutionary trends in human
populations because they do not believe
there is enough gene flow for the multiregional model to work without it. They reject
the interpretation of Templeton (1993) of
mtDNA distribution as reflecting a long history of “restricted gene flow among Old
World human populations with no single
source population for all genetic variation.”
Instead, they cite Livingstone (1992), whose
simple simulations “showed” that genes
cannot move fast enough, Stringer and
McKie (1996), who assert that the level of
gene flow that multiregional evolution requires is “improbable,” and Howell (1994, p.
304) who commented that the multiregional
mechanism “stretches the bounds of credulity . . . there is serious need for normal procedures of evolutionary biology to prevail.”
This is all wrong. Several evolutionary biologists have calculated the magnitude of
gene flow required for multiregional evolution to work, and it is very low. Estimates of
6
Although Brace’s “stages of human evolution” is very different from multiregional evolution, he treats it as an only slightly
modified version, and Smith and Harrold (1997) accept this.
7
This, of course, is not true; it is neither a valid description of
multiregional evolution, nor an accurate portrayal of its intellectual roots (see Wolpoff and Caspari, 1997).
8
However, this “probable mutation effect” is not regarded as a
valid evolutionary mechanism (Calcagno and Gibson, 1988).
134
M.H. WOLPOFF ET AL.
the number of people who need move between continents each generation for isolation by distance to explain the existing multiregional distribution of neutral genes
average only a few each generation (Harpending et al., 1996; Relethford, 1998;
Templeton, 1998). In fact, Relethford (1999)
shows that on average, less than one migrant per generation is sufficient, hardly a
large enough number to justify the comment
that multiregional evolution must require
an unacceptably large amount of migration
in order to work. The estimate of Relethford
(1999) is a maximum because it is for neutral genes. Genes under selection may have
spread with much less migration, since
their ultimate frequency and pattern of dispersion are controlled by the magnitude of
selection, which in large populations is
greater than the effect of migration on frequencies. Smith and Harrold (1997) accept
the formulation of Brace (1991), in which
culture relaxes selection and thereby creates parallel changes as structures reduce
because of mutations, as being the multiregional model. It is not, and it is not necessary to accept so improbable an explanation
as the probable mutation effect to account
for the worldwide evolution of the human
species.
DISCUSSION
“Multiregional” does not mean independent multiple origins, ancient divergence of
modern populations, simultaneous appearance of adaptive characters in different regions, or parallel evolution. By depending
on genic exchanges as the basis of its explanation of how differentiation, geographic
variation, and evolutionary change within
the human species take place, multiregional
evolution is the antithesis of these. Therefore, the incorrect portrayals and invalid
assumptions about the basis of multiregional evolution challenge our valid understanding of the issues, and undermine attempts to make progress in resolving them.
One might gather from this essay that we
contend that most authors who believe multiregional evolution is invalid, are in fact
mistaking multiregional evolution for multiple origins. To a great extent this is correct. But it is not at all universally true.
Prominent paleoanthropologists such as
Tattersall (1997) and population geneticists
such as Harpending et al. (1998) are truly
exceptions to this generalization. They describe multiregional evolution correctly, but
do not believe it is valid. There may be others who understand the model and yet disagree with it, but there are not many, and
certainly not enough to suggest that misunderstanding multiregional evolution and
the conviction it is wrong are unrelated.
Of course, the controversy does not persist solely because multiregional evolution
is improperly understood, although misunderstandings have played an important role
in how the controversy developed and their
continuation has helped this controversy
endure. Lying beneath this is a disagreement that mainly persists, and will continue
to persist, because explanations of modern
human origins through the mechanisms of
complete replacement and worldwide evolution of a single species fully contradict each
other. One of them must be wrong, and either is a valid refutation of the other. There
is no compromise position to make it easier
to find a synthesis of the views or form the
basis of a consensus that might make some
more comfortable. Instead, we would predict
that the posturing, spin-doctoring of new
data, and repositioning will continue until
one side admits it is mistaken, or more
likely, turns its attentions to other problems
and, like the Ramapithecus debate, this one
fades away. But we believe this can never
happen until accuracy is achieved in portraying the evolutionary explanation that
some authors are so sure is incorrect.
LITERATURE CITED
Awadalla P, Eyre-Walker A, Smith jm. 1999. Linkage
disequilibrium and recombination in hominid mitochondrial DNA. Science 286:2524 –2525.
Birdsell JB. 1958 On population structure in generalized hunting and collecting populations. Evolution
12:189 –205.
Bower B. 1999. DNA’s evolutionary dilemma: genetic
studies collide with the mystery of human evolution.
Sci News 155:88.
Brace CL. 1991. Cultural innovation and the mechanism for the emergence of modern morphology [abstract]. Am J Phys Anthropol [Suppl] 12:52.
Brace CL. 1995. Biocultural interaction and the mechanism of mosaic evolution. Am Anthropol 97:711–721.
Bräuer G, Stringer C. 1997. Models, polarization, and
perspectives on modern human origins. In: Clark GA,
Willermet CM, editors. Conceptual issues in modern
MULTIREGIONAL, NOT MULTIPLE ORIGINS
human origins research. New York: Aldine de
Gruyter. p 191–201, 437– 492.
Calcagno JM, Gibson KR. 1988. Human dental reduction: natural selection or the probable mutation effect. Am J Phys Anthropol 77:505–517.
Cavalli-Sforza LL. 1998. The Chinese human genome
diversity project. Proc Natl Acad Sci USA 95:11501–
11503.
Cavalli-Sforza LL, Menozzi P, Piazza A. 1993. Demic
expansions and human evolution. Science 259:639 –
646.
Chu JY, Huang W, Kuang SQ, Wang JM, Xu JJ, Chu
ZT, Yang ZQ, Lin KQ, Li P, Wu M, Geng ZC, Tan CC,
Du RF, Jin L. 1998. Genetic relationship of populations in China. Proc Natl Acad Sci USA 95:11763–
1178.
Clark GA. 1997. Through a glass darkly: conceptual
issues in modern human origins research. In: Clark
GA, Willermet CM, editors. Conceptual issues in
modern human origins research. New York: Aldine de
Gruyter. p 60 –76, 437– 492.
Clark GA, Willermet CM. 1997. Introduction to conceptual issues in modern human origins research. In:
Clark GA, Willermet CM, editors. Conceptual issues
in modern human origins research. New York: Aldine
de Gruyter. p 1– 8, 437– 492.
Duarte C, Maurı́cio J, Pettitt PB, Souto P, Trinkaus E,
van der Plicht H, Zilhão J. 1999. The early Upper
Paleolithic human skeleton from the Abrigo do Lagar
Velho (Portugal) and modern human emergence in
Iberia. Proc Natl Acad Sci USA 96:7604 –7609.
Frayer DW. 1992. The persistence of Neandertal features in post-Neandertal Europeans. In: Bräuer G,
Smith FH, editors. Continuity or replacement? Controversies in Homo sapiens evolution. Rotterdam:
Balkema. p 179 –188.
Frayer DW, Wolpoff MH, Smith FH, Thorne AG, Pope
GG. 1993. The fossil evidence for modern human origins. Am Anthropol 95:14 –50.
Gould SJ. 1994. So near and yet so far. NY Rev Books
24 –28.
Gould SL, Vrba E. 1982. Exaptation—a missing term in
the science of form. Paleobiology 8:4 –15.
Harding RM, Fullerton SM, Griffiths RC, Bond J, Cox
MJ, Schneider JA, Moulin DS, Clegg JB. 1997. Archaic African and Asian lineages in the genetic ancestry of modern humans. Am J Hum Genet 60:722–
789.
Harpending S, Batzer MA, Gurven M, Jorde LB, Rogers
AR, Sherry ST. 1998. Genetic traces of ancient demography. Proc Natl Acad Sci USA 95:1961–1967.
Harpending HC, Relethford JH, Sherry ST. 1996. Methods and models for understanding human diversity.
In: Boyce AJ, Mascie-Taylor CGN, editors. Molecular
biology and human diversity. Cambridge: Cambridge
University Press. p 283–299.
Harris EE, Hey J. 1999. X chromosome evidence for
ancient human histories. Proc Natl Acad Sci USA
96:3320 –3324.
Hey J. 1997. Mitochondrial and nuclear genes present
conflicting portraits of human origins. Mol Biol Evol
14:166 –172.
Howell FC. 1994. A chronostratigraphic and taxonomic
framework of the origins of modern humans. In: Nitecki MH, Nitecki DV, editors. Origins of anatomically modern humans. New York: Plenum Press. p
253–319.
Howell FC. 1996. Thoughts on the study and interpretation of the human fossil record. In: Meikle WE,
Howell FC, Jablonski NG, editors. Contemporary issues in human evolution. Wattis symposium series in
135
anthropology, California Academy of Sciences memoir. Volume 21. p 1– 45.
Lasker GW, Crews CE. 1996. Behavioral influences on
the evolution of human genetic diversity. Mol Biol
Evol 5:232–240.
Lewin R. 1993. The origin of modern humans. New
York: Scientific American Library.
Li J, Underhill PA, Doctor V, Davis RW, Peidong S,
Cavalli-Sforza LL, Oefner PJ. 1999. Distribution of
haplotypes from a chromosome 21 region distinguishes multiple prehistoric human migrations. Proc
Natl Acad Sci USA 96:3796 –3800.
Li T, Etler DA. 1992. New Middle Pleistocene hominid
crania from Yunxian in China. Nature 357:404 – 407.
Livingstone FB. 1992. Gene flow in the Pleistocene.
Hum Biol 64:67– 80.
Loewe L, Scherer S. 1997. Mitochondrial Eve: the plot
thickens. Tree 12:422– 423.
Lü Z. 1990. La découverte de l’homme fossile de Jingniu-shan. Prem Etud Anthropol 94:899 –902.
Moore JH. 1994. Putting anthropology back together
again: the ethnogenetic critique of cladistic theory.
Am Anthropol 96:925–948.
Nordborg M. 1998. On the probability of Neanderthal
ancestry. Am J Hum Genet 63:1237–1240.
Pope GG. 1992. Craniofacial evidence for the origin of
modern humans in China. Yrbk Phys Anthropol 35:
243–298.
Relethford JH. 1995. Genetics and modern human origins. Evol Anthropol 4:53– 63.
Relethford JH. 1998. Genetics of modern human origins
and diversity. Annu Rev Anthropol 27:1–23.
Relethford JH. 1999. Models, predictions, and the fossil
record of modern human origins. Evol Anthropol 8:7–
10.
Relethford JH, Harpending HC. 1995. Ancient differences in population size can mimic a recent African
origin of modern humans. Curr Anthropol 36:667–
674.
Seielstad M, Bekele E, Ibrahim M, Touré A, Traoré M.
1999. A view of modern human origins from Y chromosome microsatellite variation. Gene Res 9:558 –
567.
Smith FH. 1997. Modern human origins. In: Spencer F,
editor. History of physical anthropology. An encyclopedia. New York: Garland Publishing Inc. p 661– 672.
Smith SL, Harrold FB. 1997. A paradigm’s worth of
difference? Understanding the impasse over modern
human origins. Yrbk Phys Anthropol 40:113–138.
Stringer CB, McKie R. 1996. African exodus: the origins
of modern humanity. London: Jonathan Cape.
Tattersall I. 1997. Out of Africa again . . . and again?
Sci. Am 276:60 – 67.
Templeton AR. 1993. The “Eve” hypotheses: a genetic
critique and reanalysis. Am Anthropol 95:51–72.
Templeton AR. 1998. Human races: a genetic and evolutionary perspective. Am Anthropol 100 :632– 650.
Tindale NB. 1940. Distribution of Australian Aboriginal
tribes: a field survey. Trans R Soc S. Austr 64:140 –
231.
Willermet CM, Hill B. 1997. Fuzzy set theory and its
implications for species models. In: Clark GA, Willermet CM, editors. Conceptual issues in modern human
origins research. New York: Aldine de Gruyter. p
77– 88, 437– 492.
Wise CA, Sraml M, Easteal S. 1998. Departure from
neutrality at the mitochondrial NADH dehydrogenase subunit 2 gene in humans, but not in chimpanzees. Genetics 148:409 – 421.
Wise CA, Sraml M, Rubinsztein DC, Easteal S. 1997.
Comparative nuclear and mitochondrial genome di-
136
M.H. WOLPOFF ET AL.
versity in humans and chimpanzees. Mol Biol Evol
14:707–716.
Wolpoff MH. 1998. Concocting a divisive theory. Evol
Anthropol 7:1–3.
Wolpoff MH, Caspari R. 1996. An unparalleled parallelism. Anthropologie (Brno) 34:215–223.
Wolpoff MH, Caspari R. 1997. Race and human evolution. New York: Simon and Schuster.
Wolpoff MH, Wu X, Thorne AG. 1984. Modern Homo
sapiens origins: a general theory of hominid evolution
involving the fossil evidence from east Asia. In: Smith
FH, Spencer F, editors. The origins of modern humans: a world survey of the fossil evidence. New
York: Alan R. Liss. p 411– 483.
Wolpoff MH, Thorne AG, Smith FH, Frayer DW, Pope
GG. 1994. Multiregional evolution: a world-wide
source for modern human populations. In: Nitecki
MH, Nitecki DV, editors. Origins of anatomically
modern humans. New York: Plenum Press. p 175–
199.